1. Introduction
Waste management is a major issue for municipalities and governments around the world [
1]. In Canada, more than 25 million tons of municipal solid wastes (MSWs) are generated each year and have to be eliminated. This represents approximately 729 kg cap
−1 yr
−1 [
2]. One of the major challenges of landfills is the generation of polluted leachate resulting from the percolation of rainwater through the different layers of waste [
3]. A multitude of contaminants, including ammonia, dissolved organic matter, phenolic compounds, chlorides, sulfates, trace elements (TE), and xenobiotic organic compounds, can be found in leachate, which can lead to the contamination of ground and surface water near the site [
4,
5,
6,
7]. Landfill leachate shows large variability in its pollutant concentration, depending in particular on the age of the waste stored in the landfill cells [
8]. In the province of Quebec (Canada), leachate must be captured and treated before it is discharged into the surrounding environment [
9]. Conventionally, leachates are treated by expensive physico-chemical, filtration, or biological processes. The current leachate treatment practice in Quebec uses biological processes operated in alternating aerobic or anaerobic modes for carbon, ammonia, and total Kjeldahl nitrogen (TKN) removal [
10].
To overcome some of the treatment efficiency and cost challenges encountered in leachate treatment, an innovative approach was proposed and tested on operating landfill sites in Europe over the last twenty years. It consists of leachate treatment systems that combine aerated retention ponds and short-rotation coppice (SRC) using willows or poplars established on the roof of former landfill cells or on adjacent arable land irrigated with leachate [
11,
12,
13,
14]. This is a particularly appealing approach. Costs are low since the implementation and treatment both occur on-site [
15]. Furthermore, the biomass produced in various sectors of chain values (such as the production of bioenergy or other bioproducts) that are associated with the ecological services offered by such structures (such as CO
2 capture and storage, visual landscape improvement, and the increase in biodiversity) can be integrated every 3–5 years.
However, to secure these advantages, the phytofiltration systems must be functioning properly, which implies that (i) plant growth is not affected by the nature or quantities of the leachate applied and that (ii) contaminant leaching into the underground water after filtration is reduced or prevented (if on adjacent land). It has been reported that the efficiency of phytofiltration systems was subject to variations depending on plantation design setup and that symptoms of toxicity related to excessive salt concentrations (chloride ions) and nutrient imbalances (nitrogen and phosphorus) have been observed [
11,
16]. Other factors can also interfere with the effectiveness of leachate treatment, including soil texture, tree species or cultivars, the quality and quantity of the leachate applied, and irrigation methods, which can make it difficult to apply standard operating procedures in various environments [
17,
18].
Several studies showed that willows could effectively capture nitrogen, phosphorus, organic matter, and certain metals (Cd, Zn, Pb, Ni) in various applications, such as added organic fertilizers [
19,
20], municipal wastewater [
21], or leachate [
22]. More recently, Lachapelle-T et al. [
23] carried out a pilot project in southern Quebec (Canada) using willows in SRC to treat municipal wastewater and concluded that the system allowed the removal of more than 95% of the nitrogen and phosphorus contained in the effluent. More importantly, the study also demonstrated that such plantations, characterized by very high evapotranspiration rates, could receive up to 25 mL ha
−1 yr
−1 of wastewater without negative impacts on willow growth or the environment [
24]. Since landfill leachate has a high and variable contaminant concentration, little is known about the treatment ability of willows in SRC.
The objectives of this study were (1) to determine the efficacy of an SRC willow plantation when used as a vegetative filter in the pedoclimatic conditions of Quebec (Canada) for treating various loads of leachate from inoperative landfill cells and (2) to verify the impacts on soil, runoff water, and willow yields.
3. Discussion
Many studies have demonstrated that willow vegetation filters can effectively treat various sources of wastewater on-site under a diversity of climates [
25,
26,
27,
28,
29]. In this study, up to 7.5 mm d
−1 (D2/D2P treatments) of leachate was applied to willows (
Salix miyabeana ‘SX64’) grown on the clay layer topping an inoperative landfill cell. Over a cycle of 131 days, from June to October, plants were irrigated for about 110 days, thereby showing that willows are well adapted to water-saturated and compacted soils, with no negative impact on their growth and development [
12,
30,
31]. Similar large-scale studies (thousands of m
2) were carried-out with
Salix miyabeana ‘SX67’ under Quebec’s humid continental climate to treat primary municipal wastewater [
20,
22]. Cultivar ‘SX67’, a cultivar of the same willow species utilized in this study, has been shown to have a very high-water demand, allowing a high irrigation rate of up to 2961 mm yr
−1 with a former agricultural soil, which enables the treatment of up to 30,000 m
3 ha
−1 yr
−1 of municipal wastewater [
23,
32]. However, it was shown that high hydraulic loading rates (HLR) were applied to willows at the end of the season, while evapotranspiration was low and growth started reducing, causing deep percolation and nitrogen leaching into the groundwater, indicating that HLR should be modulated according to its willow filtering capability [
33].
Evapotranspiration is one of the most critical factors in large-scale industrial plant-based treatments whose main objective is to minimize the wastewater volumes to be treated conventionally [
34]. Willows can be classified among the species displaying high evapotranspiration rates [
35]. Under the climate of Eastern Canada, many willow cultivars remain photosynthetically active very late in the season, when the majority of deciduous species have lost their foliage [
36]. This is another advantage of using willows in the treatment of wastewater or leachate. In the current study, the vegetation filter was able to treat, by evapotranspiration, up to 5560 m
3 ha
−1 of leachate, exhibiting moderate electroconductivity (3.7–4.2 mS cm
−1). However, the RZDM model predicted that, over the growing season, an average of 4.5% of the total water (rainfall and irrigation) received by willows would potentially end up as runoff outside the operational filtration unit (data not shown). This water balance model also revealed root zone depletion during most of the irrigation campaign (between 10 and 70 mm for D1 and D2/D2P; also validated by in situ soil moisture tension results; data not shown), indicating that additional work would be needed to maximize the retention capability of the clay soil used in the cultivation of such landfill leachate filter systems. The relative similarity in the RZDM pattern between D1 and D2 treatments, in contrast to the D0 treatment, could be partially explained by the gain in the aboveground biomass production of leachate-irrigated willows, which promotes the evapotranspiration rate of plants as described above and, consequently, decreases soil moisture.
A significant increase in the biomass yield following leachate addition is reported here. The average biomass production in our experimental conditions after two years reached 18 tons ha
−1 for intermediate (D1) loading but exceeded values of 44 tons ha
−1 for the D2P treatment. In contrast, the biomass yield was only 17 tons ha
−1 in the groundwater treatment (D0). However, these values remain less impressive than those obtained by Jerbi et al. [
32]. In that study, carried out in a similar climatic context but on agricultural land, annual yields of 40 tons ha
−1 y
−1 were obtained when the willows were irrigated with wastewater from municipal primary effluents. It is likely that the soil conditions, as well as the various characteristics of the two effluents, explain this difference. Still, leachate had a positive effect on willow growth and induced no signs of toxicity. Apart from the water supply, fertilization is described as a driving factor that influences biomass yield and, consequently, evapotranspiration [
29]. Jerbi et al. [
32] have shown that high nitrogen inputs to the soil by wastewater irrigation resulted in a significant increase in stomatal conductance, leaf area, as well as chlorophyll and nitrogen leaf content, while the fine root:aboveground biomass ratio decreased [
32,
37]. Our data closely match these findings since leaf area and pigment content was significantly higher with the highest leachate supply (providing 1 478 kg ha
−1 NH
4-N cumulative loading over the growing period) compared to water treatment. Moreover, the leachate-irrigated willows maintained their green pigments and leaves for approximately two to three weeks longer than the control treatment at the end of the growing season, which may be part of the reason for their additional growth. It would also explain why the maximal evapotranspiration rate measured in August was two times higher in D2-treated plants (10.7 mm day
−1) than in D0-treated ones (5.5 mm day
−1). Finally, the high nitrogen supply to the soil during leachate irrigation significantly increased the growth of the longest willow stems by the end of the growing season, contributing to a higher aboveground biomass yield.
In the current study, different removal patterns were observed, with leachate contributing up to 1907 kg COD ha
−1 yr
−1 and 1478 kg NH
4-N ha
−1 yr
−1. Removal varied, on average, from 42% (D2) or 62% (D2P) to 80% (D1) for COD and was close to 100% for NH
4-N. Furthermore, the system’s filtration capacity remained effective throughout the irrigation period. The high efficiency of willow vegetation filters for removing pollutants from wastewater is well documented. A trial conducted in Sweden in 2010 showed a removal efficiency of about 80% for organic carbon (TOC) and total N by the cultivar ‘Tora’ (
S. schwerinii x S. viminalis) irrigated with leachate whose total loadings applied were 378 kg ha
−1 yr
−1 and 2021 kg ha
−1 yr
−1 for the total N and TOC, respectively [
22]. More recently, Jerbi et al. [
32] reported high removal efficiencies (93.4% for COD, 99.3% for NH
4-N, and 98.3% for P) with
S. miyabeana ‘SX67’ irrigated with municipal wastewater contributing to 8700 kg COD ha
−1 yr
−1, 1245 kg N ha
−1 yr
−1 and 121 kg P ha
−1 yr
−1. While willows can take up NH
4-N for nutritional purposes, a bacterial nitrification process could also explain its removal efficiency. The increase in NO
3–N in the superficial soil horizon was between 3.75-fold for D1 and 6.0-fold for D2/D2P at the end of the growing season. It cannot be excluded that the nitrification process could be underestimated due to the uptake by willows of this preferred nutritional nitrogen form from the soil [
38]. Interestingly, we have noticed that the NO
3–N concentration decreased by about 2–3 fold in May 2021 in leachate irrigated soil (data not shown), suggesting that a nitrification process could occur during the winter (from November 2020 to May 2021) before initiating a new irrigation cycle. Jones et al. [
18] pointed out that highly concentrated leachates (EC > 4.0 mS cm
−1) could affect the filtering capacity of the plant system, thereby limiting the volumes that could be processed (<250 m
3 ha
−1 yr
−1). This could be one of the explanations for the lower COD removal observed in the porewater of the D2 leachate treatment (EC 4.2 mS cm
−1). However, the filtering capacity of our willow filtering system was not adversely impacted by this EC range for the removal of NH
4-N, regardless of the leachate dose applied to the willows. This finding concurs with data obtained with the leachate-irrigated
S. matsudana ‘Levante’, indicating that the NH
4-N concentration significantly decreased under the lowest leachate supply (i.e., around EC 4.0 mS/cm; 29). Moreover, concentrations detected in the soil porewater after irrigation with moderately concentrated leachate (i.e., 111 mg COD L
−1 and 0.82 mg NH
4-N L
−1 for D2P treatment) did not exceed the standard limits imposed by Quebec regulations, which are 25 mg NH
3-N L
−1 and 150 mg BOD
5 L
−1 for the release of treated leachate in the environment [
39].
Phosphorus fertilization improved leachate treatment efficiency, increasing COD, TKN, and NH
4-N removal by 44%, 32%, and 64%, respectively. Soil matrix physico-chemical characteristics of the site (i.e., high clay content, low organic matter percentage, low initial phosphorus soil concentration, and soil compaction) could negatively affect the phosphorus availability of the soil and, therefore, its assimilation by plants. Furthermore, leachate added very low amounts of phosphorus to the irrigated willow filters. Consequently, one treatment (D2P) was added to the experimental design, consisting of a phosphorus amendment of 40 kg P
2O
5 ha
−1 yr
−1 when applied to willows irrigated with the highest leachate dose (D2 treatment), which corresponds to willows requirements in SRC [
40,
41,
42]. Phosphorus is essential to growth and root proliferation at the initial stages of establishment. When associated with nitrogen, it is highly effective on willow aboveground and with a root-dry biomass yield [
43]. In the present study, higher aboveground biomass yield (+25%) and leachate treatment efficiency (+44% for COD, +32% for TKN, and +64% for NH
4-N compared to D2 treatment) were achieved with P fertilization. These findings suggest that phosphorus could influence the soil bacterial community, especially species involved in degrading the organic matter and nitrogen present in the leachate. Indeed, previous research has reported phosphorus availability to be one of the major variables regulating the abundance and diversity of the bacterial community, and this shaping occurred only when phosphorus was associated with carbon or nitrogen [
44,
45,
46].
In the present study, higher amounts of nutrients were measured in leaf tissues after leachate irrigation (especially B, Fe, and N). The significant strong increase in the N content in leaf tissues of leachate-irrigated willows (+56% in D1, +86% in D2, and +76% in D2P) was likely correlated with the high amount that nitrogen leachate contributed to the soil, thus suggesting the better availability and recovery of essential nutrients by plants for their growth. The high removal efficiency of nitrogen achieved by willows has also been documented in other studies [
21,
22,
32,
47]. Nutrient interactions in crop plants are multiple and complex, acting at several physiological levels, including transportation and signaling [
48]. They may induce deficiencies or toxicities, modify growth responses, and/or modify nutrient composition [
49]. For instance, a positive interaction between phosphorus and nitrogen was reported in the context of their mutual absorption by plants and synergetic effects on plant stimulation. On the other hand, applied high levels of phosphorus may induce an iron deficiency in plants [
49]. In the context of the phytoremediation described here, a well-balanced supply of nutrients could be critical for optimizing both leachate treatment and the production of renewable high-quality aboveground wood biomass (willows are usually coppiced every three years) for commercial use.
Although the leachates used in this pilot study exhibited moderate sodium and chloride concentrations (around 300 mg L
−1) and EC, we observed that the EC of porewater increased gradually over the course of the season with leachate treatments, while it remained low and stable with the water control. This increase was faster for D2/D2P than for D1 at the beginning of the irrigation campaign, reflecting the different volumes of leachate applied for irrigation. At the end of the irrigation period, it was 3.1, 4.2, and 3.7 mS cm
−1 for D1, D2, and D2P, respectively, thus closely approaching the value measured in the leachate (data not shown). In addition, in the porewater of the D2 treatment, the total sodium reached 402 mg L
−1 at the end of the irrigation campaign. As such, there is no evidence that this salinity rate was a factor limiting the growth of willows and their efficiency in treating leachate. Nevertheless, from a long-term perspective, the progressive salinization of the soil in which the willows grow could have an antagonist effect on phytoremediation performance. Although several willow cultivars have shown good tolerance to moderate salinity treatment (7 dS m
−1), Canadian indigenous willows were reported to have a higher potential for long-term survival under severe salinity treatment (14 dS m
−1) [
50]. Moreover, recent works have reported an increased adaptation to soil salinity among tetraploid hybrids, thanks to a different Na
+ balance between the leaves and roots, which is also characteristic of the cultivar ‘SX64’ used in this study, compared to diploid ones [
51].